apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-11-03 02:53:26 +08:00
Code Issues Actions 5 Packages Projects Releases Wiki Activity

[Feature] Support_eplb (#2997)
Some checks failed
Deploy GitHub Pages / deploy (push) Has been cancelled
Details

Browse Source 
* [Feature] support_eplb

* [Feature] support_eplb

* [Fix] fix mm ep
This commit is contained in:
xiaoxiaohehe001
2025-07-24 20:22:45 +08:00
committed by GitHub
parent f37d00e856
commit 2970b00dfa
11 changed files with 118 additions and 50 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
fastdeploy/worker/experts_manager.py
View File

@@ -37,7 +37,7 @@ class RedundantExpertManger:
ep_size: int,
) -> None:
"""Initialize a redundant expert manager"""
self.num_expert = n_routed_experts
self.num_expert = n_routed_experts if isinstance(n_routed_experts, int) else n_routed_experts[0]
self.redundant_experts_num = redundant_experts_num
self.num_hidden_layers = num_hidden_layers